With the rapid development of broadband services, the interconnection between networks is used more and more and bears more services. A protection can be made on links and nodes thereon by adopting various methods inside a network according to the technology the network uses. With an increasingly stronger protection demand for traffics and its higher and higher requirements, certain operators also propose a protection demand required for network junction areas. A way of link aggregation can be used for implementing the protection. The common way can be the link aggregation and it can also be the loop protection. The ring network protection technology is limited by network topology, which is not applicable to the network junction parts which are in a fully connected network way. In the existing standard 802.1AX-link aggregation of the Institute of Electrical and Electronic Engineers (IEEE), a protection of multiple links can be achieved, but it is limited to support the link aggregation of a single node, which cannot satisfy a demand for node protection.
In order to adapt to a higher diversification of networking ways in inter-network junction areas and to implement the protection of links and edge nodes, the IEEE standard organization proposes to implement Distributed Resilient Network Interconnect (DRNI) by extending the 802.1AX-link aggregation, so as to support the multi-node link aggregation. The 802.1AX is extended by the DRNI, which makes that each end of a link aggregation group can be a logical node constituted by one or more nodes for the aggregation, that is, a distributed Link Aggregation Group (LAG) is constituted by multiple links of these nodes, thereby guaranteeing a dual protection for links and nodes of the network junction areas.
As shown in FIG. 1, both ends of the LAG are respectively composed of one portal (Portal), each portal consists of two systems, wherein, a portal 1 (Portal1) is composed of a system A and a system B, the system A and system B are called “intra-end systems” (called “system” for short) of the Portal1, the system A and the system B are systems adjacent to each other. A portal 2 (Portal2) is composed of a system C and a system D. A link 1, link 2, link 3 and link 4 between the Portal1 and the Portal2 are aggregated into a distributed LAG.
At present, a distribution way according to service messages is used in the DRNI, when a node receives a message, the service message is forwarded according to an identifier (e.g., a Virtual Local Area Network Identifier (VLAN ID, called VID for short)) of the service message. When the node receives a message required to be sent to the interior of the current network or a message required to be sent to a Network Node Interface (NNI) side via an aggregated link of the current node, it is required to judge whether a service identifier of the message is in a port service table and an aggregator service table, if the VID is in the port service table, the message is forwarded, and if the VID is not in the port service table, the message is discarded. When a failure occurs in the link or node, it is required to hand over the service to other ports or other nodes to send, but there is no such service at other ports or nodes, which causes that the service is discarded and traffic is interrupted.
With respect to this situation, there is no solution at present about how to ensure that fast service handover is implemented by renegotiating to select an aggregated link between the nodes when a failure occurs in the link or node so as to reduce the influence of the failure on the service.